COMPGROUPS.NET | Search | Post Question | Groups | Stream | About | Register

### Within-factors ANOVA

• Email
• Follow

```I have three independent variables in a within-subjects experiment:

1) Size (Small/Large)
2) Disp (0/100/200)
3) Items (4/8)

So, in total I have 12 test conditions. For each test condition, I'm
taking multiple trials and the number of trials is unequal.

When Items = 4, I take 10 trials.   (6 test conditions)
When items = 8, I take 20 trials.   (6 test conditions)

Now, what type of ANOVA should I perform on this data? I'm a first
time SPSS user, so I'm unclear how to enter all the data in SPSS?
Mainly, I'm confused how to enter data for different trials
corresponding to the same test condition.
```
 0

See related articles to this posting

```Sorry, I wrote within-factors ANOVA instead of Within-subjects ANOVA

On Mar 27, 6:40=A0pm, Akhil <akhilmath...@gmail.com> wrote:
> I have three independent variables in a within-subjects experiment:
>
> 1) Size (Small/Large)
> 2) Disp (0/100/200)
> 3) Items (4/8)
>
> So, in total I have 12 test conditions. For each test condition, I'm
> taking multiple trials and the number of trials is unequal.
>
> When Items =3D 4, I take 10 trials. =A0 (6 test conditions)
> When items =3D 8, I take 20 trials. =A0 (6 test conditions)
>
> Now, what type of ANOVA should I perform on this data? I'm a first
> time SPSS user, so I'm unclear how to enter all the data in SPSS?
> Mainly, I'm confused how to enter data for different trials
> corresponding to the same test condition.

```
 0

```Akhil wrote:
> I have three independent variables in a within-subjects experiment:
>
> 1) Size (Small/Large)
> 2) Disp (0/100/200)
> 3) Items (4/8)
>
> So, in total I have 12 test conditions. For each test condition, I'm
> taking multiple trials and the number of trials is unequal.
>
> When Items = 4, I take 10 trials.   (6 test conditions)
> When items = 8, I take 20 trials.   (6 test conditions)
>
> Now, what type of ANOVA should I perform on this data? I'm a first
> time SPSS user, so I'm unclear how to enter all the data in SPSS?
> Mainly, I'm confused how to enter data for different trials
> corresponding to the same test condition.

Are you computing some kind of summary measure over trials (e.g.,
a mean or median), or is Trial another factor in the ANOVA model?

--
Bruce Weaver
"When all else fails, RTFM."
```
 0

```On Mar 28, 11:23=A0am, Bruce Weaver <bwea...@lakeheadu.ca> wrote:
> Akhil wrote:
> > I have three independent variables in a within-subjects experiment:
>
> > 1) Size (Small/Large)
> > 2) Disp (0/100/200)
> > 3) Items (4/8)
>
> > So, in total I have 12 test conditions. For each test condition, I'm
> > taking multiple trials and the number of trials is unequal.
>
> > When Items =3D 4, I take 10 trials. =A0 (6 test conditions)
> > When items =3D 8, I take 20 trials. =A0 (6 test conditions)
>
> > Now, what type of ANOVA should I perform on this data? I'm a first
> > time SPSS user, so I'm unclear how to enter all the data in SPSS?
> > Mainly, I'm confused how to enter data for different trials
> > corresponding to the same test condition.
>
> Are you computing some kind of summary measure over trials (e.g.,
> a mean or median), or is Trial another factor in the ANOVA model?
>
> --
> Bruce Weaver
> "When all else fails, RTFM."

Thanks for the reply. Yes, I'm calculating a summary measure (mean)
over trials.
```
 0

```On Sun, 28 Mar 2010 13:33:05 -0700 (PDT), Akhil
<akhilmathurs@gmail.com> wrote:

>On Mar 28, 11:23�am, Bruce Weaver <bwea...@lakeheadu.ca> wrote:
>> Akhil wrote:
>> > I have three independent variables in a within-subjects experiment:
>>
>> > 1) Size (Small/Large)
>> > 2) Disp (0/100/200)
>> > 3) Items (4/8)
>>
>> > So, in total I have 12 test conditions. For each test condition, I'm
>> > taking multiple trials and the number of trials is unequal.
>>
>> > When Items = 4, I take 10 trials. � (6 test conditions)
>> > When items = 8, I take 20 trials. � (6 test conditions)
>>
>> > Now, what type of ANOVA should I perform on this data? I'm a first
>> > time SPSS user, so I'm unclear how to enter all the data in SPSS?
>> > Mainly, I'm confused how to enter data for different trials
>> > corresponding to the same test condition.
>>
>> Are you computing some kind of summary measure over trials (e.g.,
>> a mean or median), or is Trial another factor in the ANOVA model?
>>
>> --
>> Bruce Weaver
>> "When all else fails, RTFM."
>
>Thanks for the reply. Yes, I'm calculating a summary measure (mean)
>over trials.

Using the summary says that your file, going into analysis,
can conveniently have just ID plus 12 variables - systematically
varying size, disp, and items.  Repeated Measures is most
conveniently performed with all the Repeats in one record.

For convenience and accuracy of data entry, assuming that you
want all of the data in SPSS, I recommend that you use one line
to enter each of the 12 conditions, and list the 10 or 20 scores on
that line.  This will be be relatively easy to proofread, etc., to
avoid data entry errors.

Then you can use MEAN(v1 to v20)  to get the mean for all
conditions, and use VARSTOCASES  to create one line for
each ID.

--
Rich Ulrich
```
 0

```On Mar 28, 5:45=A0pm, Rich Ulrich <rich.ulr...@comcast.net> wrote:
> On Sun, 28 Mar 2010 13:33:05 -0700 (PDT), Akhil
>
>
>
> <akhilmath...@gmail.com> wrote:
> >On Mar 28, 11:23=A0am, Bruce Weaver <bwea...@lakeheadu.ca> wrote:
> >> Akhil wrote:
> >> > I have three independent variables in a within-subjects experiment:
>
> >> > 1) Size (Small/Large)
> >> > 2) Disp (0/100/200)
> >> > 3) Items (4/8)
>
> >> > So, in total I have 12 test conditions. For each test condition, I'm
> >> > taking multiple trials and the number of trials is unequal.
>
> >> > When Items =3D 4, I take 10 trials. =A0 (6 test conditions)
> >> > When items =3D 8, I take 20 trials. =A0 (6 test conditions)
>
> >> > Now, what type of ANOVA should I perform on this data? I'm a first
> >> > time SPSS user, so I'm unclear how to enter all the data in SPSS?
> >> > Mainly, I'm confused how to enter data for different trials
> >> > corresponding to the same test condition.
>
> >> Are you computing some kind of summary measure over trials (e.g.,
> >> a mean or median), or is Trial another factor in the ANOVA model?
>
> >> --
> >> Bruce Weaver
> >> "When all else fails, RTFM."
>
> >Thanks for the reply. Yes, I'm calculating a summary measure (mean)
> >over trials.
>
> Using the summary says that your file, going into analysis,
> can conveniently have just ID plus 12 variables - systematically
> varying size, disp, and items. =A0Repeated Measures is most
> conveniently performed with all the Repeats in one record.
>
> For convenience and accuracy of data entry, assuming that you
> want all of the data in SPSS, I recommend that you use one line
> to enter each of the 12 conditions, and list the 10 or 20 scores on
> that line. =A0This will be be relatively easy to proofread, etc., to
> avoid data entry errors.
>
> Then you can use MEAN(v1 to v20) =A0to get the mean for all
> conditions, and use VARSTOCASES =A0to create one line for
> each ID.
>
> --
> Rich Ulrich

I'll just give a bit more detail on Rich's suggestion, in case it's
not clear.  He is suggesting entering the data into a file that has
the following variables:

ID - a unique ID for each subject
Size (1=3DSmall, 2=3DLarge)
Disp (0/100/200)
Items (4/8)
V1 - V20:  variables to hold the scores from the 10 or 20 trials

Be sure to save the file after entering the raw data.

After computing a mean (of V1 to V20) on each line, use VARSTOCASES
with Size, Disp and Items as index variables.  The raw scores are not
needed any longer, so can be excluded via /DROP.   This will result in
a file that has one row per ID, with 12 variables for your 12
combinations of the 3 factors.

NOTE:  Be careful not to overwrite the first file with this second
one.

--
Bruce Weaver
"When all else fails, RTFM."
```
 0

5 Replies
549 Views

Similar Articles

12/9/2013 4:22:19 PM
page loaded in 340996 ms. (1)

Similar Artilces:

Re: Factor Analysis on Factors
charles19 <charles19@SYMPATICO.CA> wrote >Sent: Jun 26, 2008 9:43 AM >To: SAS-L@LISTSERV.UGA.EDU >Subject: Factor Analysis on Factors > >Hi Everyone: > >Good to join this group. We're a fun bunch. :-) Nice to have you I am doing exploratory factor analysis on >several measures related to human personality development. The >resulting factors are interesting. I am wondering if it is possible to >do a factor analysis on the factors themselves? In this way, I hope to >cluster the factors into "mega factors" (Neurosis-Passivity; >... to this list, and someone (maybe even me) is sure to be able to help. As you probably know, the literature on personality is vast, and I know that some people have already done this sort of analysis --- but the last time I looked at this was about 10 years ago. >I don't know much about confirmatory factor analysis. Would the latter >yield this type of information (i.e., higher level factors)? Thanks >for your help, > Unusually for statistics, in this case "confirmatory" means more or less what it means in ordinary English. :-). You use CFA when you already have

Linux TCO Factors
Linux TCO Factors Demir Barlas April 12 2004 Linux, the open source operating system, is often associated with a lower total cost of ownership (TCO) than its proprietary rivals. Interestingly, however, research group Meta had taken the position that Linux had the same TCO as Windows. A new alert from the company revises if not reverses this position, stating that, in Meta's opinion, Linux TCO can indeed be driven down 10 to 40 percent ... http://www.line56.com/articles/default.asp?ArticleID=5539

MC: All pairs of factors of an integer
Mini-challenge for RPL enthusiasts by Joe Horn It's been forever since the last "mini-challenge". So here's a really easy one that everybody can try. Write the smallest RPL program that gives all the *pairs* of factors of an integer. Example: Input: 12 Output: { { 1 12 } { 2 6 } { 3 4 } } Another example: Input: 25 Output: { { 1 25 } { 5 5 } } Full description: Write an RPL program that inputs any integer and outputs every possible *pair* of integer factors. The output is a list of lists, with each sublist containing a pair of factors, as in the example above. Output must be complete. The order of the factors in each pair, and the order of the pairs, does not matter. Repeats are not allowed. The usual mini-challenge rules apply. Only native User-RPL is allowed. Embedding code into a string then executing it via OBJ-> or STR-> is a Bad Thing. And so on. The SMALLEST program (in bytes) wins. Prize: eternal fame in the comp.sys.hp48 archives. -Joe- http://holyjoe.net/hhc2009 In article <d9455264-975c-4243-82ce-60bc7f9248f9@v15g2000prn.googlegroups.com>, Joe Horn <joehorn@holyjoe.net> wrote: > Mini-challenge for RPL

Deciding factors to move from 32-bit to 64-bit application??
Currently, my application runs as a 32-bit application. I am wondering whether it will help in improving the performance, if I make it 64-bit. What are the factors that I should consider, while deciding about it? ["Followup-To:" header set to comp.unix.programmer.] On 27 Sep 2006 07:05:19 -0700, qazmlp1209@rediffmail.com <qazmlp1209@rediffmail.com> wrote: > Currently, my application runs as a 32-bit application. I am wondering > whether it will help in improving the performance, if I make it 64-bit. > What are the factors that I should consider, while deciding... 64-bit unless you need the VM, but you might want to have a go at it on amd64. -- Mikko Rauhala - mjr@iki.fi - <URL:http://www.iki.fi/mjr/> Transhumanist - WTA member - <URL:http://www.transhumanism.org/> Singularitarian - SIAI supporter - <URL:http://www.singinst.org/> Mikko Rauhala wrote: > So, on sparc, don't go 64-bit unless you need the VM, but you might > want to have a go at it on amd64. There are ancillary factors - for instance, 32-bit processes have limits on open file descriptors and select. A 64-bit version will have much higher limits